Showing 1–50 of 150 results for author: Pan, R

GUIDE: Towards Scalable Advising for Research Ideas

Authors: Yaowenqi Liu, BingXu Meng, Rui Pan, Jerry Huang, Tong Zhang

Abstract: The field of AI research is advancing at an unprecedented pace, enabling automated hypothesis generation and experimental design across diverse domains such as biology, mathematics, and artificial intelligence. Despite these advancements, there remains a significant gap in the availability of scalable advising systems capable of providing high-quality, well-reasoned feedback to refine proposed hyp… ▽ More The field of AI research is advancing at an unprecedented pace, enabling automated hypothesis generation and experimental design across diverse domains such as biology, mathematics, and artificial intelligence. Despite these advancements, there remains a significant gap in the availability of scalable advising systems capable of providing high-quality, well-reasoned feedback to refine proposed hypotheses and experimental designs. To address this challenge, we explore key factors that underlie the development of robust advising systems, including model size, context length, confidence estimation, and structured reasoning processes. Our findings reveal that a relatively small model, when equipped with a well-compressed literature database and a structured reasoning framework, can outperform powerful general-purpose language models such as Deepseek-R1 in terms of acceptance rates for self-ranked top-30% submissions to ICLR 2025. Moreover, when limited to high-confidence predictions, our system achieves an acceptance rate exceeding 90% on the ICLR 2025 test set, underscoring its potential to significantly enhance the quality and efficiency of hypothesis generation and experimental design. The code is released at https://github.com/HowardLiu0830/GUIDE-Research-Idea-Evaluation. △ Less

Submitted 9 July, 2025; originally announced July 2025.

Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities

Authors: Gheorghe Comanici, Eric Bieber, Mike Schaekermann, Ice Pasupat, Noveen Sachdeva, Inderjit Dhillon, Marcel Blistein, Ori Ram, Dan Zhang, Evan Rosen, Luke Marris, Sam Petulla, Colin Gaffney, Asaf Aharoni, Nathan Lintz, Tiago Cardal Pais, Henrik Jacobsson, Idan Szpektor, Nan-Jiang Jiang, Krishna Haridasan, Ahmed Omran, Nikunj Saunshi, Dara Bahri, Gaurav Mishra, Eric Chu , et al. (3284 additional authors not shown)

Abstract: In this report, we introduce the Gemini 2.X model family: Gemini 2.5 Pro and Gemini 2.5 Flash, as well as our earlier Gemini 2.0 Flash and Flash-Lite models. Gemini 2.5 Pro is our most capable model yet, achieving SoTA performance on frontier coding and reasoning benchmarks. In addition to its incredible coding and reasoning skills, Gemini 2.5 Pro is a thinking model that excels at multimodal unde… ▽ More In this report, we introduce the Gemini 2.X model family: Gemini 2.5 Pro and Gemini 2.5 Flash, as well as our earlier Gemini 2.0 Flash and Flash-Lite models. Gemini 2.5 Pro is our most capable model yet, achieving SoTA performance on frontier coding and reasoning benchmarks. In addition to its incredible coding and reasoning skills, Gemini 2.5 Pro is a thinking model that excels at multimodal understanding and it is now able to process up to 3 hours of video content. Its unique combination of long context, multimodal and reasoning capabilities can be combined to unlock new agentic workflows. Gemini 2.5 Flash provides excellent reasoning abilities at a fraction of the compute and latency requirements and Gemini 2.0 Flash and Flash-Lite provide high performance at low latency and cost. Taken together, the Gemini 2.X model generation spans the full Pareto frontier of model capability vs cost, allowing users to explore the boundaries of what is possible with complex agentic problem solving. △ Less

Submitted 22 July, 2025; v1 submitted 7 July, 2025; originally announced July 2025.

Comments: 72 pages, 17 figures

Chain-of-Experts: Unlocking the Communication Power of Mixture-of-Experts Models

Authors: Zihan Wang, Rui Pan, Jiarui Yao, Robert Csordas, Linjie Li, Lu Yin, Jiajun Wu, Tong Zhang, Manling Li, Shiwei Liu

Abstract: We propose Chain-of-Experts (CoE), a new Mixture-of-Experts (MoE) architecture that introduces sequential expert communication within each layer. Unlike traditional MoE models, where experts operate independently in parallel, CoE processes tokens iteratively across a chain of experts inside a layer. To support dynamic expert selection across iterations, CoE employs a dedicated router at each itera… ▽ More We propose Chain-of-Experts (CoE), a new Mixture-of-Experts (MoE) architecture that introduces sequential expert communication within each layer. Unlike traditional MoE models, where experts operate independently in parallel, CoE processes tokens iteratively across a chain of experts inside a layer. To support dynamic expert selection across iterations, CoE employs a dedicated router at each iteration step within a layer. This design allows tokens to re-evaluate and select different experts during each iteration, rather than being statically assigned. As a result, CoE introduces a flexible routing mechanism that increases the diversity of expert combinations and enriches the model's representational capacity. CoE demonstrates improved performance under fixed compute: on math reasoning tasks, it reduces validation loss from 1.20 to 1.12 compared to a standard MoE. Beyond performance, CoE offers a new scaling axis: depth through expert iteration, which complements conventional width/depth scaling. For example, using 2x iterations matches the performance of 3x expert selections (in width), while reducing memory usage by 17.6-42% relative to other scaling strategies. Our analysis reveals that CoE's benefits stem from its iterative residual structure and enhanced expert specialization empowered by iterative routing, which together unlock more expressive representations. Code is available at https://github.com/ZihanWang314/coe. △ Less

Submitted 22 June, 2025; originally announced June 2025.

VL-GenRM: Enhancing Vision-Language Verification via Vision Experts and Iterative Training

Authors: Jipeng Zhang, Kehao Miao, Renjie Pi, Zhaowei Wang, Runtao Liu, Rui Pan, Tong Zhang

Abstract: Reinforcement Fine-Tuning (RFT) with verifiable rewards has advanced large language models but remains underexplored for Vision-Language (VL) models. The Vision-Language Reward Model (VL-RM) is key to aligning VL models by providing structured feedback, yet training effective VL-RMs faces two major challenges. First, the bootstrapping dilemma arises as high-quality training data depends on already… ▽ More Reinforcement Fine-Tuning (RFT) with verifiable rewards has advanced large language models but remains underexplored for Vision-Language (VL) models. The Vision-Language Reward Model (VL-RM) is key to aligning VL models by providing structured feedback, yet training effective VL-RMs faces two major challenges. First, the bootstrapping dilemma arises as high-quality training data depends on already strong VL models, creating a cycle where self-generated supervision reinforces existing biases. Second, modality bias and negative example amplification occur when VL models hallucinate incorrect visual attributes, leading to flawed preference data that further misguides training. To address these issues, we propose an iterative training framework leveraging vision experts, Chain-of-Thought (CoT) rationales, and Margin-based Rejection Sampling. Our approach refines preference datasets, enhances structured critiques, and iteratively improves reasoning. Experiments across VL-RM benchmarks demonstrate superior performance in hallucination detection and multimodal reasoning, advancing VL model alignment with reinforcement learning. △ Less

Submitted 16 June, 2025; originally announced June 2025.

Subjective Perspectives within Learned Representations Predict High-Impact Innovation

Authors: Likun Cao, Rui Pan, James Evans

Abstract: Existing studies of innovation emphasize the power of social structures to shape innovation capacity. Emerging machine learning approaches, however, enable us to model innovators' personal perspectives and interpersonal innovation opportunities as a function of their prior trajectories of experience. We theorize then quantify subjective perspectives and innovation opportunities based on innovator… ▽ More Existing studies of innovation emphasize the power of social structures to shape innovation capacity. Emerging machine learning approaches, however, enable us to model innovators' personal perspectives and interpersonal innovation opportunities as a function of their prior trajectories of experience. We theorize then quantify subjective perspectives and innovation opportunities based on innovator positions within the geometric space of concepts inscribed by dynamic language representations. Using data on millions of scientists, inventors, writers, entrepreneurs, and Wikipedia contributors across the creative domains of science, technology, film, entrepreneurship, and Wikipedia, here we show that measured subjective perspectives anticipate what ideas individuals and groups creatively attend to and successfully combine in future. When perspective and background diversity are decomposed as the angular difference between collaborators' perspectives on their creation and between their experiences, the former consistently anticipates creative achievement while the latter portends its opposite, across all cases and time periods examined. We analyze a natural experiment and simulate creative collaborations between AI (large language model) agents designed with various perspective and background diversity, which are consistent with our observational findings. We explore mechanisms underlying these findings and identify how successful collaborators leverage common language to weave together diverse experience obtained through trajectories of prior work that converge to provoke one another and innovate. We explore the importance of these findings for team assembly and research policy. △ Less

Submitted 5 June, 2025; originally announced June 2025.

Comments: 107 pages, 20 figures

MINT: Memory-Infused Prompt Tuning at Test-time for CLIP

Authors: Jiaming Yi, Ruirui Pan, Jishen Yang, Xiulong Yang

Abstract: Improving the generalization ability of Vision-Language Pre-trained Models (VLMs) under test-time data distribution shifts remains a critical challenge. The existing Test-Time Adaptation (TTA) methods fall short in fully leveraging the model's internal knowledge, particularly in dynamically adapting to complex and hierarchical visual semantic information. In this paper, we propose Memory-Infused P… ▽ More Improving the generalization ability of Vision-Language Pre-trained Models (VLMs) under test-time data distribution shifts remains a critical challenge. The existing Test-Time Adaptation (TTA) methods fall short in fully leveraging the model's internal knowledge, particularly in dynamically adapting to complex and hierarchical visual semantic information. In this paper, we propose Memory-Infused Prompt Tuning (MINT), a novel framework to address this issue. Inspired by human associative memory theory, MINT introduces a Memory Prompt Bank (MPB), which stores learnable key-value prompt pairs that work as a memory of previously seen samples. During the test time, relevant prompt pairs in the MPB are retrieved by the hierarchical visual features of test images to dynamically assemble Associative Prompts. The associative prompts are then injected into the image encoder for fine-grained, customized visual contextual guidance. MINT also utilizes learnable text prompts. MINT thus enables rapid, precise VLM adaptation at test time by leveraging this MPB-acquired memory, without source data or retraining. The code is available at https://github.com/Jamieyi2004/MINT. △ Less

Submitted 31 May, 2025; originally announced June 2025.

Comments: 14 pages, 3 figures

Understanding Overadaptation in Supervised Fine-Tuning: The Role of Ensemble Methods

Authors: Yifan Hao, Xingyuan Pan, Hanning Zhang, Chenlu Ye, Rui Pan, Tong Zhang

Abstract: Supervised fine-tuning (SFT) on domain-specific data is the dominant approach for adapting foundation models to specialized tasks. However, it has been observed that SFT models tend to forget knowledge acquired during pretraining. In vision models, ensembling a pretrained model with its fine-tuned counterpart has been shown to mitigate this issue. In this work, we demonstrate that the same holds f… ▽ More Supervised fine-tuning (SFT) on domain-specific data is the dominant approach for adapting foundation models to specialized tasks. However, it has been observed that SFT models tend to forget knowledge acquired during pretraining. In vision models, ensembling a pretrained model with its fine-tuned counterpart has been shown to mitigate this issue. In this work, we demonstrate that the same holds for language models, and, more strikingly, we observe an overadaptation phenomenon: the ensemble model not only retains general knowledge from the foundation model but also outperforms the fine-tuned model even on the fine-tuning domain itself. Despite the empirical success of ensembling, a theoretical understanding of its benefits remains underexplored. We develop a formal theoretical analysis of the overadaptation phenomenon. Ensembling mitigates this by balancing two primary sources of error: bias, caused by insufficient fine-tuning, and variance, introduced by overfitting to fine-tuning data. While regularization techniques aim to address this trade-off, we show that ensembling provides a more effective solution. We analyze this phenomenon in over-parameterized linear settings and demonstrate that interpolating between pretrained and fine-tuned weights significantly improves performance. These findings offer theoretical justification for the observed advantages of model ensembling, supported by empirical experiments consistent with our analysis. △ Less

Submitted 2 June, 2025; originally announced June 2025.

Structured Gradient Guidance for Few-Shot Adaptation in Large Language Models

Authors: Hongye Zheng, Yichen Wang, Ray Pan, Guiran Liu, Binrong Zhu, Hanlu Zhang

Abstract: This paper presents a gradient-informed fine-tuning method for large language models under few-shot conditions. The goal is to enhance task adaptability and training stability when data is limited. The method builds on a base loss function and introduces two gradient-related regularization terms. The first enforces gradient direction consistency to guide parameter updates along task-relevant direc… ▽ More This paper presents a gradient-informed fine-tuning method for large language models under few-shot conditions. The goal is to enhance task adaptability and training stability when data is limited. The method builds on a base loss function and introduces two gradient-related regularization terms. The first enforces gradient direction consistency to guide parameter updates along task-relevant directions and prevent drift. The second controls gradient magnitude to avoid abnormal updates. Together, these components support a more efficient and stable optimization path. To further improve cross-task generalization, the method incorporates a gradient alignment mechanism. This mechanism measures the consistency between optimization directions of the source and target tasks. It enhances fine-tuning performance in multi-task and cross-domain scenarios. Across various natural language understanding tasks, the method outperforms existing fine-tuning strategies in average accuracy, gradient stability, and directional alignment. Empirical evaluations under different sample sizes and domain-specific tasks confirm the method's robustness and broad applicability in low-resource environments. In particular, the method shows clear advantages in controlling parameter update paths. The results demonstrate that a gradient-based fine-tuning framework can effectively leverage the representational power of large language models. It ensures training stability while reducing dependence on large volumes of labeled data. △ Less

Submitted 31 May, 2025; originally announced June 2025.

MiCRo: Mixture Modeling and Context-aware Routing for Personalized Preference Learning

Authors: Jingyan Shen, Jiarui Yao, Rui Yang, Yifan Sun, Feng Luo, Rui Pan, Tong Zhang, Han Zhao

Abstract: Reward modeling is a key step in building safe foundation models when applying reinforcement learning from human feedback (RLHF) to align Large Language Models (LLMs). However, reward modeling based on the Bradley-Terry (BT) model assumes a global reward function, failing to capture the inherently diverse and heterogeneous human preferences. Hence, such oversimplification limits LLMs from supporti… ▽ More Reward modeling is a key step in building safe foundation models when applying reinforcement learning from human feedback (RLHF) to align Large Language Models (LLMs). However, reward modeling based on the Bradley-Terry (BT) model assumes a global reward function, failing to capture the inherently diverse and heterogeneous human preferences. Hence, such oversimplification limits LLMs from supporting personalization and pluralistic alignment. Theoretically, we show that when human preferences follow a mixture distribution of diverse subgroups, a single BT model has an irreducible error. While existing solutions, such as multi-objective learning with fine-grained annotations, help address this issue, they are costly and constrained by predefined attributes, failing to fully capture the richness of human values. In this work, we introduce MiCRo, a two-stage framework that enhances personalized preference learning by leveraging large-scale binary preference datasets without requiring explicit fine-grained annotations. In the first stage, MiCRo introduces context-aware mixture modeling approach to capture diverse human preferences. In the second stage, MiCRo integrates an online routing strategy that dynamically adapts mixture weights based on specific context to resolve ambiguity, allowing for efficient and scalable preference adaptation with minimal additional supervision. Experiments on multiple preference datasets demonstrate that MiCRo effectively captures diverse human preferences and significantly improves downstream personalization. △ Less

Submitted 30 May, 2025; originally announced May 2025.

Active Learning for Multiple Change Point Detection in Non-stationary Time Series with Deep Gaussian Processes

Authors: Hao Zhao, Rong Pan

Abstract: Multiple change point (MCP) detection in non-stationary time series is challenging due to the variety of underlying patterns. To address these challenges, we propose a novel algorithm that integrates Active Learning (AL) with Deep Gaussian Processes (DGPs) for robust MCP detection. Our method leverages spectral analysis to identify potential changes and employs AL to strategically select new sampl… ▽ More Multiple change point (MCP) detection in non-stationary time series is challenging due to the variety of underlying patterns. To address these challenges, we propose a novel algorithm that integrates Active Learning (AL) with Deep Gaussian Processes (DGPs) for robust MCP detection. Our method leverages spectral analysis to identify potential changes and employs AL to strategically select new sampling points for improved efficiency. By incorporating the modeling flexibility of DGPs with the change-identification capabilities of spectral methods, our approach adapts to diverse spectral change behaviors and effectively localizes multiple change points. Experiments on both simulated and real-world data demonstrate that our method outperforms existing techniques in terms of detection accuracy and sampling efficiency for non-stationary time series. △ Less

Submitted 26 May, 2025; originally announced May 2025.

Requirements Coverage-Guided Minimization for Natural Language Test Cases

Authors: Rongqi Pan, Feifei Niu, Lionel C. Briand, Hanyang Hu

Abstract: As software systems evolve, test suites tend to grow in size and often contain redundant test cases. Such redundancy increases testing effort, time, and cost. Test suite minimization (TSM) aims to eliminate such redundancy while preserving key properties such as requirement coverage and fault detection capability. In this paper, we propose RTM (Requirement coverage-guided Test suite Minimization),… ▽ More As software systems evolve, test suites tend to grow in size and often contain redundant test cases. Such redundancy increases testing effort, time, and cost. Test suite minimization (TSM) aims to eliminate such redundancy while preserving key properties such as requirement coverage and fault detection capability. In this paper, we propose RTM (Requirement coverage-guided Test suite Minimization), a novel TSM approach designed for requirement-based testing (validation), which can effectively reduce test suite redundancy while ensuring full requirement coverage and a high fault detection rate (FDR) under a fixed minimization budget. Based on common practice in critical systems where functional safety is important, we assume test cases are specified in natural language and traced to requirements before being implemented. RTM preprocesses test cases using three different preprocessing methods, and then converts them into vector representations using seven text embedding techniques. Similarity values between vectors are computed utilizing three distance functions. A Genetic Algorithm, whose population is initialized by coverage-preserving initialization strategies, is then employed to identify an optimized subset containing diverse test cases matching the set budget. We evaluate RTM on an industrial automotive system dataset comprising $736$ system test cases and $54$ requirements. Experimental results show that RTM consistently outperforms baseline techniques in terms of FDR across different minimization budgets while maintaining full requirement coverage. Furthermore, we investigate the impact of test suite redundancy levels on the effectiveness of TSM, providing new insights into optimizing requirement-based test suites under practical constraints. △ Less

Submitted 26 May, 2025; originally announced May 2025.

AstroMLab 4: Benchmark-Topping Performance in Astronomy Q&A with a 70B-Parameter Domain-Specialized Reasoning Model

Authors: Tijmen de Haan, Yuan-Sen Ting, Tirthankar Ghosal, Tuan Dung Nguyen, Alberto Accomazzi, Emily Herron, Vanessa Lama, Rui Pan, Azton Wells, Nesar Ramachandra

Abstract: General-purpose large language models, despite their broad capabilities, often struggle with specialized domain knowledge, a limitation particularly pronounced in more accessible, lower-parameter versions. This gap hinders their deployment as effective agents in demanding fields such as astronomy. Building on our prior work with AstroSage-8B, this study introduces AstroSage-70B, a significantly la… ▽ More General-purpose large language models, despite their broad capabilities, often struggle with specialized domain knowledge, a limitation particularly pronounced in more accessible, lower-parameter versions. This gap hinders their deployment as effective agents in demanding fields such as astronomy. Building on our prior work with AstroSage-8B, this study introduces AstroSage-70B, a significantly larger and more advanced domain-specialized natural-language AI assistant. It is designed for research and education across astronomy, astrophysics, space science, astroparticle physics, cosmology, and astronomical instrumentation. Developed from the Llama-3.1-70B foundation, AstroSage-70B underwent extensive continued pre-training on a vast corpus of astronomical literature, followed by supervised fine-tuning and model merging. Beyond its 70-billion parameter scale, this model incorporates refined datasets, judiciously chosen learning hyperparameters, and improved training procedures, achieving state-of-the-art performance on complex astronomical tasks. Notably, we integrated reasoning chains into the SFT dataset, enabling AstroSage-70B to either answer the user query immediately, or first emit a human-readable thought process. Evaluated on the AstroMLab-1 benchmark -- comprising 4,425 questions from literature withheld during training -- AstroSage-70B achieves state-of-the-art performance. It surpasses all other tested open-weight and proprietary models, including leading systems like o3, Gemini-2.5-Pro, Claude-3.7-Sonnet, Deepseek-R1, and Qwen-3-235B, even those with API costs two orders of magnitude higher. This work demonstrates that domain specialization, when applied to large-scale models, can enable them to outperform generalist counterparts in specialized knowledge areas like astronomy, thereby advancing the frontier of AI capabilities in the field. △ Less

Submitted 23 May, 2025; originally announced May 2025.

TVR: Automotive System Requirement Traceability Validation and Recovery Through Retrieval-Augmented Generation

Authors: Feifei Niu, Rongqi Pan, Lionel C. Briand, Hanyang Hu, Krishna Koravadi

Abstract: In automotive software development, as well as other domains, traceability between stakeholder requirements and system requirements is crucial to ensure consistency, correctness, and regulatory compliance. However, erroneous or missing traceability relationships often arise due to improper propagation of requirement changes or human errors in requirement mapping, leading to inconsistencies and inc… ▽ More In automotive software development, as well as other domains, traceability between stakeholder requirements and system requirements is crucial to ensure consistency, correctness, and regulatory compliance. However, erroneous or missing traceability relationships often arise due to improper propagation of requirement changes or human errors in requirement mapping, leading to inconsistencies and increased maintenance costs. Existing approaches do not address traceability between stakeholder and system requirements, rely on open-source data -- as opposed to automotive (or any industry) data -- and do not address the validation of manual links established by engineers. Additionally, automotive requirements often exhibit variations in the way they are expressed, posing challenges for supervised models requiring training. The recent advancements in large language models (LLMs) provide new opportunities to address these challenges. In this paper, we introduce TVR, a requirement Traceability Validation and Recovery approach primarily targeting automotive systems, leveraging LLMs enhanced with retrieval-augmented generation (RAG). TVR is designed to validate existing traceability links and recover missing ones with high accuracy. We empirically evaluate TVR on automotive requirements, achieving 98.87% accuracy in traceability validation and 85.50% correctness in traceability recovery. Additionally, TVR demonstrates strong robustness, achieving 97.13% in accuracy when handling unseen requirements variations. The results highlight the practical effectiveness of RAG-based LLM approaches in industrial settings, offering a promising solution for improving requirements traceability in complex automotive systems. △ Less

Submitted 15 June, 2025; v1 submitted 21 April, 2025; originally announced April 2025.

SpecReason: Fast and Accurate Inference-Time Compute via Speculative Reasoning

Authors: Rui Pan, Yinwei Dai, Zhihao Zhang, Gabriele Oliaro, Zhihao Jia, Ravi Netravali

Abstract: Recent advances in inference-time compute have significantly improved performance on complex tasks by generating long chains of thought (CoTs) using Large Reasoning Models (LRMs). However, this improved accuracy comes at the cost of high inference latency due to the length of generated reasoning sequences and the autoregressive nature of decoding. Our key insight in tackling these overheads is tha… ▽ More Recent advances in inference-time compute have significantly improved performance on complex tasks by generating long chains of thought (CoTs) using Large Reasoning Models (LRMs). However, this improved accuracy comes at the cost of high inference latency due to the length of generated reasoning sequences and the autoregressive nature of decoding. Our key insight in tackling these overheads is that LRM inference, and the reasoning that it embeds, is highly tolerant of approximations: complex tasks are typically broken down into simpler steps, each of which brings utility based on the semantic insight it provides for downstream steps rather than the exact tokens it generates. Accordingly, we introduce SpecReason, a system that automatically accelerates LRM inference by using a lightweight model to (speculatively) carry out simpler intermediate reasoning steps and reserving the costly base model only to assess (and potentially correct) the speculated outputs. Importantly, SpecReason's focus on exploiting the semantic flexibility of thinking tokens in preserving final-answer accuracy is complementary to prior speculation techniques, most notably speculative decoding, which demands token-level equivalence at each step. Across a variety of reasoning benchmarks, SpecReason achieves $1.4-3.0\times$ speedup over vanilla LRM inference while improving accuracy by $0.4-9.0\%$. Compared to speculative decoding without SpecReason, their combination yields an additional $8.8-58.0\%$ latency reduction. We open-source SpecReason at https://github.com/ruipeterpan/specreason. △ Less

Submitted 16 May, 2025; v1 submitted 10 April, 2025; originally announced April 2025.

Beyond LLMs: A Linguistic Approach to Causal Graph Generation from Narrative Texts

Authors: Zehan Li, Ruhua Pan, Xinyu Pi

Abstract: We propose a novel framework for generating causal graphs from narrative texts, bridging high-level causality and detailed event-specific relationships. Our method first extracts concise, agent-centered vertices using large language model (LLM)-based summarization. We introduce an "Expert Index," comprising seven linguistically informed features, integrated into a Situation-Task-Action-Consequence… ▽ More We propose a novel framework for generating causal graphs from narrative texts, bridging high-level causality and detailed event-specific relationships. Our method first extracts concise, agent-centered vertices using large language model (LLM)-based summarization. We introduce an "Expert Index," comprising seven linguistically informed features, integrated into a Situation-Task-Action-Consequence (STAC) classification model. This hybrid system, combining RoBERTa embeddings with the Expert Index, achieves superior precision in causal link identification compared to pure LLM-based approaches. Finally, a structured five-iteration prompting process refines and constructs connected causal graphs. Experiments on 100 narrative chapters and short stories demonstrate that our approach consistently outperforms GPT-4o and Claude 3.5 in causal graph quality, while maintaining readability. The open-source tool provides an interpretable, efficient solution for capturing nuanced causal chains in narratives. △ Less

Submitted 10 April, 2025; originally announced April 2025.

Comments: published at the 7th Workshop on Narrative Understanding, NAACL 2025

AdaCoder: An Adaptive Planning and Multi-Agent Framework for Function-Level Code Generation

Authors: Yueheng Zhu, Chao Liu, Xuan He, Xiaoxue Ren, Zhongxin Liu, Ruwei Pan, Hongyu Zhang

Abstract: Recently, researchers have proposed many multi-agent frameworks for function-level code generation, which aim to improve software development productivity by automatically generating function-level source code based on task descriptions. A typical multi-agent framework consists of Large Language Model (LLM)-based agents that are responsible for task planning, code generation, testing, debugging, e… ▽ More Recently, researchers have proposed many multi-agent frameworks for function-level code generation, which aim to improve software development productivity by automatically generating function-level source code based on task descriptions. A typical multi-agent framework consists of Large Language Model (LLM)-based agents that are responsible for task planning, code generation, testing, debugging, etc. Studies have shown that existing multi-agent code generation frameworks perform well on ChatGPT. However, their generalizability across other foundation LLMs remains unexplored systematically. In this paper, we report an empirical study on the generalizability of four state-of-the-art multi-agent code generation frameworks across six open-source LLMs with varying parameter sizes, architectures, and performance levels. Our study reveals the unstable generalizability of existing frameworks on diverse foundation LLMs. Based on the findings obtained from the empirical study, we propose AdaCoder, a novel adaptive planning, multi-agent framework for function-level code generation. AdaCoder has two phases. Phase-1 is an initial code generation step without planning, which uses an LLM-based coding agent and a script-based testing agent to unleash LLM's native power, identify cases beyond LLM's power, and determine the errors hindering execution. Phase-2 adds a rule-based debugging agent and an LLM-based planning agent for iterative code generation with planning. Our evaluation shows that AdaCoder achieves higher generalizability on diverse LLMs. Compared to the best baseline MapCoder, AdaCoder is on average 27.69% higher in Pass@1, 16 times faster in inference, and 12 times lower in token consumption. △ Less

Submitted 5 April, 2025; originally announced April 2025.

From Observation to Orientation: an Adaptive Integer Programming Approach to Intervention Design

Authors: Abdelmonem Elrefaey, Rong Pan

Abstract: Using both observational and experimental data, a causal discovery process can identify the causal relationships between variables. A unique adaptive intervention design paradigm is presented in this work, where causal directed acyclic graphs (DAGs) are for effectively recovered with practical budgetary considerations. In order to choose treatments that optimize information gain under these consid… ▽ More Using both observational and experimental data, a causal discovery process can identify the causal relationships between variables. A unique adaptive intervention design paradigm is presented in this work, where causal directed acyclic graphs (DAGs) are for effectively recovered with practical budgetary considerations. In order to choose treatments that optimize information gain under these considerations, an iterative integer programming (IP) approach is proposed, which drastically reduces the number of experiments required. Simulations over a broad range of graph sizes and edge densities are used to assess the effectiveness of the suggested approach. Results show that the proposed adaptive IP approach achieves full causal graph recovery with fewer intervention iterations and variable manipulations than random intervention baselines, and it is also flexible enough to accommodate a variety of practical constraints. △ Less

Submitted 9 May, 2025; v1 submitted 3 April, 2025; originally announced April 2025.

ASGO: Adaptive Structured Gradient Optimization

Authors: Kang An, Yuxing Liu, Rui Pan, Yi Ren, Shiqian Ma, Donald Goldfarb, Tong Zhang

Abstract: Training deep neural networks is a structured optimization problem, because the parameters are naturally represented by matrices and tensors rather than by vectors. Under this structural representation, it has been widely observed that gradients are low-rank and Hessians are approximately block-wise diagonal. These structured properties are crucial for designing efficient optimization algorithms,… ▽ More Training deep neural networks is a structured optimization problem, because the parameters are naturally represented by matrices and tensors rather than by vectors. Under this structural representation, it has been widely observed that gradients are low-rank and Hessians are approximately block-wise diagonal. These structured properties are crucial for designing efficient optimization algorithms, but are not utilized by many current popular optimizers like Adam. In this paper, we present a novel optimization algorithm ASGO that capitalizes on these properties by employing a preconditioner that is adaptively updated using structured gradients. By fine-grained theoretical analysis, ASGO is proven to achieve superior convergence rates compared to existing structured gradient methods. Based on the convergence theory, we further demonstrate that ASGO can benefit from the low-rank and block-wise diagonal properties. We also discuss practical modifications of ASGO and empirically verify ASGO's effectiveness on language model tasks. △ Less

Submitted 22 June, 2025; v1 submitted 26 March, 2025; originally announced March 2025.

Comments: 30 pages

Gemma 3 Technical Report

Authors: Gemma Team, Aishwarya Kamath, Johan Ferret, Shreya Pathak, Nino Vieillard, Ramona Merhej, Sarah Perrin, Tatiana Matejovicova, Alexandre Ramé, Morgane Rivière, Louis Rouillard, Thomas Mesnard, Geoffrey Cideron, Jean-bastien Grill, Sabela Ramos, Edouard Yvinec, Michelle Casbon, Etienne Pot, Ivo Penchev, Gaël Liu, Francesco Visin, Kathleen Kenealy, Lucas Beyer, Xiaohai Zhai, Anton Tsitsulin , et al. (191 additional authors not shown)

Abstract: We introduce Gemma 3, a multimodal addition to the Gemma family of lightweight open models, ranging in scale from 1 to 27 billion parameters. This version introduces vision understanding abilities, a wider coverage of languages and longer context - at least 128K tokens. We also change the architecture of the model to reduce the KV-cache memory that tends to explode with long context. This is achie… ▽ More We introduce Gemma 3, a multimodal addition to the Gemma family of lightweight open models, ranging in scale from 1 to 27 billion parameters. This version introduces vision understanding abilities, a wider coverage of languages and longer context - at least 128K tokens. We also change the architecture of the model to reduce the KV-cache memory that tends to explode with long context. This is achieved by increasing the ratio of local to global attention layers, and keeping the span on local attention short. The Gemma 3 models are trained with distillation and achieve superior performance to Gemma 2 for both pre-trained and instruction finetuned versions. In particular, our novel post-training recipe significantly improves the math, chat, instruction-following and multilingual abilities, making Gemma3-4B-IT competitive with Gemma2-27B-IT and Gemma3-27B-IT comparable to Gemini-1.5-Pro across benchmarks. We release all our models to the community. △ Less

Submitted 25 March, 2025; originally announced March 2025.

Safe RLHF-V: Safe Reinforcement Learning from Multi-modal Human Feedback

Authors: Jiaming Ji, Xinyu Chen, Rui Pan, Conghui Zhang, Han Zhu, Jiahao Li, Donghai Hong, Boyuan Chen, Jiayi Zhou, Kaile Wang, Juntao Dai, Chi-Min Chan, Yida Tang, Sirui Han, Yike Guo, Yaodong Yang

Abstract: Multimodal large language models (MLLMs) are essential for building general-purpose AI assistants; however, they pose increasing safety risks. How can we ensure safety alignment of MLLMs to prevent undesired behaviors? Going further, it is critical to explore how to fine-tune MLLMs to preserve capabilities while meeting safety constraints. Fundamentally, this challenge can be formulated as a min-m… ▽ More Multimodal large language models (MLLMs) are essential for building general-purpose AI assistants; however, they pose increasing safety risks. How can we ensure safety alignment of MLLMs to prevent undesired behaviors? Going further, it is critical to explore how to fine-tune MLLMs to preserve capabilities while meeting safety constraints. Fundamentally, this challenge can be formulated as a min-max optimization problem. However, existing datasets have not yet disentangled single preference signals into explicit safety constraints, hindering systematic investigation in this direction. Moreover, it remains an open question whether such constraints can be effectively incorporated into the optimization process for multi-modal models. In this work, we present the first exploration of the Safe RLHF-V -- the first multimodal safety alignment framework. The framework consists of: $\mathbf{(I)}$ BeaverTails-V, the first open-source dataset featuring dual preference annotations for helpfulness and safety, supplemented with multi-level safety labels (minor, moderate, severe); $\mathbf{(II)}$ Beaver-Guard-V, a multi-level guardrail system to proactively defend against unsafe queries and adversarial attacks. Applying the guard model over five rounds of filtering and regeneration significantly enhances the precursor model's overall safety by an average of 40.9%. $\mathbf{(III)}$ Based on dual preference, we initiate the first exploration of multi-modal safety alignment within a constrained optimization. Experimental results demonstrate that Safe RLHF effectively improves both model helpfulness and safety. Specifically, Safe RLHF-V enhances model safety by 34.2% and helpfulness by 34.3%. △ Less

Submitted 22 May, 2025; v1 submitted 22 March, 2025; originally announced March 2025.

Modularization is Better: Effective Code Generation with Modular Prompting

Authors: Ruwei Pan, Hongyu Zhang

Abstract: Large Language Models are transforming software development by automatically generating code. Current prompting techniques such as Chain-of-Thought (CoT) suggest tasks step by step and the reasoning process follows a linear structure, which hampers the understanding of complex programming problems, particularly those requiring hierarchical solutions. Inspired by the principle of modularization in… ▽ More Large Language Models are transforming software development by automatically generating code. Current prompting techniques such as Chain-of-Thought (CoT) suggest tasks step by step and the reasoning process follows a linear structure, which hampers the understanding of complex programming problems, particularly those requiring hierarchical solutions. Inspired by the principle of modularization in software development, in this work, we propose a novel prompting technique, called MoT, to enhance the code generation performance of LLMs. At first, MoT exploits modularization principles to decompose complex programming problems into smaller, independent reasoning steps, enabling a more structured and interpretable problem-solving process. This hierarchical structure improves the LLM's ability to comprehend complex programming problems. Then, it structures the reasoning process using an MLR Graph (Multi-Level Reasoning Graph), which hierarchically organizes reasoning steps. This approach enhances modular understanding and ensures better alignment between reasoning steps and the generated code, significantly improving code generation performance. Our experiments on two advanced LLMs (GPT-4o-mini and DeepSeek-R1), comparing MoT to six baseline prompting techniques across six widely used datasets, HumanEval, HumanEval-ET, HumanEval+, MBPP, MBPP-ET, and MBPP+, demonstrate that MoT significantly outperforms existing baselines (e.g., CoT and SCoT), achieving Pass@1 scores ranging from 58.1% to 95.1%. The experimental results confirm that MoT significantly enhances the performance of LLM-based code generation. △ Less

Submitted 16 March, 2025; originally announced March 2025.

AgentDroid: A Multi-Agent Framework for Detecting Fraudulent Android Applications

Authors: Ruwei Pan, Hongyu Zhang, Zhonghao Jiang, Ran Hou

Abstract: With the increasing prevalence of fraudulent Android applications such as fake and malicious applications, it is crucial to detect them with high accuracy and adaptability. This paper introduces AgentDroid, a novel framework for Android fraudulent application detection based on multi-modal analysis and multi-agent systems. AgentDroid overcomes the limitations of traditional detection methods such… ▽ More With the increasing prevalence of fraudulent Android applications such as fake and malicious applications, it is crucial to detect them with high accuracy and adaptability. This paper introduces AgentDroid, a novel framework for Android fraudulent application detection based on multi-modal analysis and multi-agent systems. AgentDroid overcomes the limitations of traditional detection methods such as the inability to handle multimodal data and high false alarm rates. It processes Android applications and extracts a series of multi-modal data for analysis. Multiple LLM-based agents with specialized roles analyze the relevant data and collaborate to detect complex fraud effectively. We constructed a dataset containing various categories of fraudulent applications and legitimate applications and validated our framework on this dataset. Experimental results indicate that our multi-agent framework based on GPT-4o achieves an accuracy of 91.7% and an F1-Score of 91.68%, showing improved detection accuracy over the baseline methods. △ Less

Submitted 15 March, 2025; originally announced March 2025.

MA-LoT: Model-Collaboration Lean-based Long Chain-of-Thought Reasoning enhances Formal Theorem Proving

Authors: Ruida Wang, Rui Pan, Yuxin Li, Jipeng Zhang, Yizhen Jia, Shizhe Diao, Renjie Pi, Junjie Hu, Tong Zhang

Abstract: Solving mathematical problems using computer-verifiable languages like Lean has significantly impacted the mathematical and computer science communities. State-of-the-art methods utilize a single Large Language Model (LLM) to generate complete proof or perform tree search, but they fail to balance these tasks. We propose **MA-LoT**: *Model-CollAboration Lean-based Long Chain-of-Thought*, a compreh… ▽ More Solving mathematical problems using computer-verifiable languages like Lean has significantly impacted the mathematical and computer science communities. State-of-the-art methods utilize a single Large Language Model (LLM) to generate complete proof or perform tree search, but they fail to balance these tasks. We propose **MA-LoT**: *Model-CollAboration Lean-based Long Chain-of-Thought*, a comprehensive framework for Lean4 theorem proving to solve this issue. It separates the cognition tasks of general NL for whole-proof generation and error analysis for proof correction using the model-collaboration method. We achieve this by structured interaction of the LLM and Lean4 verifier in Long CoT. To implement the framework, we propose the novel *LoT-Transfer Learning* training-inference pipeline, which enables the Long CoT thinking capability to LLMs without special data annotation. Extensive experiment shows that our framework achieves a **61.07%** accuracy rate on the Lean4 version of the MiniF2F-Test dataset, largely outperforming DeepSeek-V3 (33.61%), single-model tree search (InternLM-Step-Prover, 50.70%), and whole-proof generation (Godel-Prover, 55.33%) baselines. Furthermore, our findings highlight the potential of combining Long CoT with formal verification for a more insightful generation in a broader perspective. △ Less

Submitted 27 May, 2025; v1 submitted 5 March, 2025; originally announced March 2025.

Ariadne: A Hotness-Aware and Size-Adaptive Compressed Swap Technique for Fast Application Relaunch and Reduced CPU Usage on Mobile Devices

Authors: Yu Liang, Aofeng Shen, Chun Jason Xue, Riwei Pan, Haiyu Mao, Nika Mansouri Ghiasi, Qingcai Jiang, Rakesh Nadig, Lei Li, Rachata Ausavarungnirun, Mohammad Sadrosadati, Onur Mutlu

Abstract: Growing application memory demands and concurrent usage are making mobile device memory scarce. When memory pressure is high, current mobile systems use a RAM-based compressed swap scheme (called ZRAM) to compress unused execution-related data (called anonymous data in Linux) in main memory. We observe that the state-of-the-art ZRAM scheme prolongs relaunch latency and wastes CPU time because it… ▽ More Growing application memory demands and concurrent usage are making mobile device memory scarce. When memory pressure is high, current mobile systems use a RAM-based compressed swap scheme (called ZRAM) to compress unused execution-related data (called anonymous data in Linux) in main memory. We observe that the state-of-the-art ZRAM scheme prolongs relaunch latency and wastes CPU time because it does not differentiate between hot and cold data or leverage different compression chunk sizes and data locality. We make three new observations. 1) anonymous data has different levels of hotness. Hot data, used during application relaunch, is usually similar between consecutive relaunches. 2) when compressing the same amount of anonymous data, small-size compression is very fast, while large-size compression achieves a better compression ratio. 3) there is locality in data access during application relaunch. We propose Ariadne, a compressed swap scheme for mobile devices that reduces relaunch latency and CPU usage with three key techniques. 1) a low-overhead hotness-aware data organization scheme aims to quickly identify the hotness of anonymous data without significant overhead. 2) a size-adaptive compression scheme uses different compression chunk sizes based on the data's hotness level to ensure fast decompression of hot and warm data. 3) a proactive decompression scheme predicts the next set of data to be used and decompresses it in advance, reducing the impact of data swapping back into main memory during application relaunch. Our experimental evaluation results on Google Pixel 7 show that, on average, Ariadne reduces application relaunch latency by 50% and decreases the CPU usage of compression and decompression procedures by 15% compared to the state-of-the-art ZRAM scheme. △ Less

Submitted 18 February, 2025; originally announced February 2025.

Comments: This is an extended version of a paper that will appear in HPCA 2025

Otter: Generating Tests from Issues to Validate SWE Patches

Authors: Toufique Ahmed, Jatin Ganhotra, Rangeet Pan, Avraham Shinnar, Saurabh Sinha, Martin Hirzel

Abstract: While there has been plenty of work on generating tests from existing code, there has been limited work on generating tests from issues. A correct test must validate the code patch that resolves the issue. This paper focuses on the scenario where that code patch does not yet exist. Doing so supports two major use-cases. First, it supports TDD (test-driven development), the discipline of "test firs… ▽ More While there has been plenty of work on generating tests from existing code, there has been limited work on generating tests from issues. A correct test must validate the code patch that resolves the issue. This paper focuses on the scenario where that code patch does not yet exist. Doing so supports two major use-cases. First, it supports TDD (test-driven development), the discipline of "test first, write code later" that has well-documented benefits for human software engineers. Second, it also validates SWE (software engineering) agents, which generate code patches for resolving issues. This paper introduces TDD-Bench-Verified, a benchmark for generating tests from issues, and Otter, an LLM-based solution for this task. Otter augments LLMs with rule-based analysis to check and repair their outputs, and introduces a novel self-reflective action planner. Experiments show Otter outperforming state-of-the-art systems for generating tests from issues, in addition to enhancing systems that generate patches from issues. We hope that Otter helps make developers more productive at resolving issues and leads to more robust, well-tested code. △ Less

Submitted 30 May, 2025; v1 submitted 7 February, 2025; originally announced February 2025.

Comments: Accepted to the main technical track of the International Conference on Machine Learning (ICML), 2025

Adapt-Pruner: Adaptive Structural Pruning for Efficient Small Language Model Training

Authors: Rui Pan, Boyao Wang, Shizhe Diao, Xingyuan Pan, Jipeng Zhang, Renjie Pi, Tong Zhang

Abstract: Small language models (SLMs) have attracted considerable attention from both academia and industry due to their broad range of applications in edge devices. To obtain SLMs with strong performance, conventional approaches either pre-train the models from scratch, which incurs substantial computational costs, or compress/prune existing large language models (LLMs), which results in performance drops… ▽ More Small language models (SLMs) have attracted considerable attention from both academia and industry due to their broad range of applications in edge devices. To obtain SLMs with strong performance, conventional approaches either pre-train the models from scratch, which incurs substantial computational costs, or compress/prune existing large language models (LLMs), which results in performance drops and falls short in comparison to pre-training. In this paper, we investigate the family of acceleration methods that involve both structured pruning and model training. We found 1) layer-wise adaptive pruning (Adapt-Pruner) is extremely effective in LLMs and yields significant improvements over existing pruning techniques, 2) adaptive pruning equipped with further training leads to models comparable to those pre-training from scratch, 3) incremental pruning brings non-trivial performance gain by interleaving pruning with training and only removing a small portion of neurons ($\sim$5%) at a time. Experimental results on LLaMA-3.1-8B demonstrate that Adapt-Pruner outperforms conventional pruning methods, such as LLM-Pruner, FLAP, and SliceGPT, by an average of 1%-7% in accuracy on commonsense benchmarks. Additionally, Adapt-Pruner restores the performance of MobileLLM-125M to 600M on the MMLU benchmark with 200$\times$ fewer tokens via pruning from its larger counterparts, and discovers a new 1B model that surpasses LLaMA-3.2-1B in multiple benchmarks. The official code is released at https://github.com/research4pan/AdaptPruner. △ Less

Submitted 14 June, 2025; v1 submitted 5 February, 2025; originally announced February 2025.

Humanity's Last Exam

Authors: Long Phan, Alice Gatti, Ziwen Han, Nathaniel Li, Josephina Hu, Hugh Zhang, Chen Bo Calvin Zhang, Mohamed Shaaban, John Ling, Sean Shi, Michael Choi, Anish Agrawal, Arnav Chopra, Adam Khoja, Ryan Kim, Richard Ren, Jason Hausenloy, Oliver Zhang, Mantas Mazeika, Dmitry Dodonov, Tung Nguyen, Jaeho Lee, Daron Anderson, Mikhail Doroshenko, Alun Cennyth Stokes , et al. (1084 additional authors not shown)

Abstract: Benchmarks are important tools for tracking the rapid advancements in large language model (LLM) capabilities. However, benchmarks are not keeping pace in difficulty: LLMs now achieve over 90\% accuracy on popular benchmarks like MMLU, limiting informed measurement of state-of-the-art LLM capabilities. In response, we introduce Humanity's Last Exam (HLE), a multi-modal benchmark at the frontier of… ▽ More Benchmarks are important tools for tracking the rapid advancements in large language model (LLM) capabilities. However, benchmarks are not keeping pace in difficulty: LLMs now achieve over 90\% accuracy on popular benchmarks like MMLU, limiting informed measurement of state-of-the-art LLM capabilities. In response, we introduce Humanity's Last Exam (HLE), a multi-modal benchmark at the frontier of human knowledge, designed to be the final closed-ended academic benchmark of its kind with broad subject coverage. HLE consists of 2,500 questions across dozens of subjects, including mathematics, humanities, and the natural sciences. HLE is developed globally by subject-matter experts and consists of multiple-choice and short-answer questions suitable for automated grading. Each question has a known solution that is unambiguous and easily verifiable, but cannot be quickly answered via internet retrieval. State-of-the-art LLMs demonstrate low accuracy and calibration on HLE, highlighting a significant gap between current LLM capabilities and the expert human frontier on closed-ended academic questions. To inform research and policymaking upon a clear understanding of model capabilities, we publicly release HLE at https://lastexam.ai. △ Less

Submitted 19 April, 2025; v1 submitted 24 January, 2025; originally announced January 2025.

Comments: 29 pages, 6 figures

DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning

Authors: DeepSeek-AI, Daya Guo, Dejian Yang, Haowei Zhang, Junxiao Song, Ruoyu Zhang, Runxin Xu, Qihao Zhu, Shirong Ma, Peiyi Wang, Xiao Bi, Xiaokang Zhang, Xingkai Yu, Yu Wu, Z. F. Wu, Zhibin Gou, Zhihong Shao, Zhuoshu Li, Ziyi Gao, Aixin Liu, Bing Xue, Bingxuan Wang, Bochao Wu, Bei Feng, Chengda Lu , et al. (175 additional authors not shown)

Abstract: We introduce our first-generation reasoning models, DeepSeek-R1-Zero and DeepSeek-R1. DeepSeek-R1-Zero, a model trained via large-scale reinforcement learning (RL) without supervised fine-tuning (SFT) as a preliminary step, demonstrates remarkable reasoning capabilities. Through RL, DeepSeek-R1-Zero naturally emerges with numerous powerful and intriguing reasoning behaviors. However, it encounters… ▽ More We introduce our first-generation reasoning models, DeepSeek-R1-Zero and DeepSeek-R1. DeepSeek-R1-Zero, a model trained via large-scale reinforcement learning (RL) without supervised fine-tuning (SFT) as a preliminary step, demonstrates remarkable reasoning capabilities. Through RL, DeepSeek-R1-Zero naturally emerges with numerous powerful and intriguing reasoning behaviors. However, it encounters challenges such as poor readability, and language mixing. To address these issues and further enhance reasoning performance, we introduce DeepSeek-R1, which incorporates multi-stage training and cold-start data before RL. DeepSeek-R1 achieves performance comparable to OpenAI-o1-1217 on reasoning tasks. To support the research community, we open-source DeepSeek-R1-Zero, DeepSeek-R1, and six dense models (1.5B, 7B, 8B, 14B, 32B, 70B) distilled from DeepSeek-R1 based on Qwen and Llama. △ Less

Submitted 22 January, 2025; originally announced January 2025.

CodeCoR: An LLM-Based Self-Reflective Multi-Agent Framework for Code Generation

Authors: Ruwei Pan, Hongyu Zhang, Chao Liu

Abstract: Code generation aims to produce code that fulfills requirements written in natural languages automatically. Large language Models (LLMs) like ChatGPT have demonstrated promising effectiveness in this area. Nonetheless, these LLMs often fail to ensure the syntactic and semantic correctness of the generated code. Recently, researchers proposed multi-agent frameworks that guide LLMs with different pr… ▽ More Code generation aims to produce code that fulfills requirements written in natural languages automatically. Large language Models (LLMs) like ChatGPT have demonstrated promising effectiveness in this area. Nonetheless, these LLMs often fail to ensure the syntactic and semantic correctness of the generated code. Recently, researchers proposed multi-agent frameworks that guide LLMs with different prompts to analyze programming tasks, generate code, perform testing in a sequential workflow. However, the performance of the workflow is not robust as the code generation depends on the performance of each agent. To address this challenge, we propose CodeCoR, a self-reflective multi-agent framework that evaluates the effectiveness of each agent and their collaborations. Specifically, for a given task description, four agents in CodeCoR generate prompts, code, test cases, and repair advice, respectively. Each agent generates more than one output and prunes away the low-quality ones. The generated code is tested in the local environment: the code that fails to pass the generated test cases is sent to the repair agent and the coding agent re-generates the code based on repair advice. Finally, the code that passes the most number of generated test cases is returned to users. Our experiments on four widely used datasets, HumanEval, HumanEval-ET, MBPP, and MBPP-ET, demonstrate that CodeCoR significantly outperforms existing baselines (e.g., CodeCoT and MapCoder), achieving an average Pass@1 score of 77.8%. △ Less

Submitted 13 January, 2025; originally announced January 2025.

Distilling Desired Comments for Enhanced Code Review with Large Language Models

Authors: Yongda Yu, Lei Zhang, Guoping Rong, Haifeng Shen, Jiahao Zhang, Haoxiang Yan, Guohao Shi, Dong Shao, Ruiqi Pan, Yuan Li, Qiushi Wang, Zhao Tian

Abstract: There has been a growing interest in using Large Language Models (LLMs) for code review thanks to their proven proficiency in code comprehension. The primary objective of most review scenarios is to generate desired review comments (DRCs) that explicitly identify issues to trigger code fixes. However, existing LLM-based solutions are not so effective in generating DRCs for various reasons such as… ▽ More There has been a growing interest in using Large Language Models (LLMs) for code review thanks to their proven proficiency in code comprehension. The primary objective of most review scenarios is to generate desired review comments (DRCs) that explicitly identify issues to trigger code fixes. However, existing LLM-based solutions are not so effective in generating DRCs for various reasons such as hallucination. To enhance their code review ability, they need to be fine-tuned with a customized dataset that is ideally full of DRCs. Nevertheless, such a dataset is not yet available, while manual annotation of DRCs is too laborious to be practical. In this paper, we propose a dataset distillation method, Desiview, which can automatically construct a distilled dataset by identifying DRCs from a code review dataset. Experiments on the CodeReviewer dataset comprising more than 150K review entries show that Desiview achieves an impressive performance of 88.93%, 80.37%, 86.67%, and 84.44% in terms of Precision, Recall, Accuracy, and F1, respectively, surpassing state-of-the-art methods. To validate the effect of such a distilled dataset on enhancing LLMs' code review ability, we first fine-tune the latest LLaMA series (i.e., LLaMA 3 and LLaMA 3.1) to build model Desiview4FT. We then enhance the model training effect through KTO alignment by feeding those review comments identified as non-DRCs to the LLMs, resulting in model Desiview4FA. Verification results indicate that Desiview4FA slightly outperforms Desiview4FT, while both models have significantly improved against the base models in terms of generating DRCs. Human evaluation confirms that both models identify issues more accurately and tend to generate review comments that better describe the issues contained in the code than the base LLMs do. △ Less

Submitted 5 January, 2025; v1 submitted 28 December, 2024; originally announced December 2024.

Comments: 12 pages, 9 figures

ACM Class: D.2.3; I.2.7

DeepSeek-V3 Technical Report

Authors: DeepSeek-AI, Aixin Liu, Bei Feng, Bing Xue, Bingxuan Wang, Bochao Wu, Chengda Lu, Chenggang Zhao, Chengqi Deng, Chenyu Zhang, Chong Ruan, Damai Dai, Daya Guo, Dejian Yang, Deli Chen, Dongjie Ji, Erhang Li, Fangyun Lin, Fucong Dai, Fuli Luo, Guangbo Hao, Guanting Chen, Guowei Li, H. Zhang, Han Bao , et al. (175 additional authors not shown)

Abstract: We present DeepSeek-V3, a strong Mixture-of-Experts (MoE) language model with 671B total parameters with 37B activated for each token. To achieve efficient inference and cost-effective training, DeepSeek-V3 adopts Multi-head Latent Attention (MLA) and DeepSeekMoE architectures, which were thoroughly validated in DeepSeek-V2. Furthermore, DeepSeek-V3 pioneers an auxiliary-loss-free strategy for loa… ▽ More We present DeepSeek-V3, a strong Mixture-of-Experts (MoE) language model with 671B total parameters with 37B activated for each token. To achieve efficient inference and cost-effective training, DeepSeek-V3 adopts Multi-head Latent Attention (MLA) and DeepSeekMoE architectures, which were thoroughly validated in DeepSeek-V2. Furthermore, DeepSeek-V3 pioneers an auxiliary-loss-free strategy for load balancing and sets a multi-token prediction training objective for stronger performance. We pre-train DeepSeek-V3 on 14.8 trillion diverse and high-quality tokens, followed by Supervised Fine-Tuning and Reinforcement Learning stages to fully harness its capabilities. Comprehensive evaluations reveal that DeepSeek-V3 outperforms other open-source models and achieves performance comparable to leading closed-source models. Despite its excellent performance, DeepSeek-V3 requires only 2.788M H800 GPU hours for its full training. In addition, its training process is remarkably stable. Throughout the entire training process, we did not experience any irrecoverable loss spikes or perform any rollbacks. The model checkpoints are available at https://github.com/deepseek-ai/DeepSeek-V3. △ Less

Submitted 18 February, 2025; v1 submitted 26 December, 2024; originally announced December 2024.

Align Anything: Training All-Modality Models to Follow Instructions with Language Feedback

Authors: Jiaming Ji, Jiayi Zhou, Hantao Lou, Boyuan Chen, Donghai Hong, Xuyao Wang, Wenqi Chen, Kaile Wang, Rui Pan, Jiahao Li, Mohan Wang, Josef Dai, Tianyi Qiu, Hua Xu, Dong Li, Weipeng Chen, Jun Song, Bo Zheng, Yaodong Yang

Abstract: Reinforcement learning from human feedback (RLHF) has proven effective in enhancing the instruction-following capabilities of large language models; however, it remains underexplored in the cross-modality domain. As the number of modalities increases, aligning all-modality models with human intentions -- such as instruction following -- becomes a pressing challenge. In this work, we make the first… ▽ More Reinforcement learning from human feedback (RLHF) has proven effective in enhancing the instruction-following capabilities of large language models; however, it remains underexplored in the cross-modality domain. As the number of modalities increases, aligning all-modality models with human intentions -- such as instruction following -- becomes a pressing challenge. In this work, we make the first attempt to fine-tune all-modality models (i.e. input and output with any modality, also named any-to-any models) using human preference data across all modalities (including text, image, audio, and video), ensuring its behavior aligns with human intentions. This endeavor presents several challenges. First, there is no large-scale all-modality human preference data in existing open-source resources, as most datasets are limited to specific modalities, predominantly text and image. Secondly, the effectiveness of binary preferences in RLHF for post-training alignment in complex all-modality scenarios remains an unexplored area. Finally, there is a lack of a systematic framework to evaluate the capabilities of all-modality models, particularly regarding modality selection and synergy. To address these challenges, we propose the align-anything framework, which includes meticulously annotated 200k all-modality human preference data. Then, we introduce an alignment method that learns from unified language feedback, effectively capturing complex modality-specific human preferences and enhancing the model's instruction-following capabilities. Furthermore, to assess performance improvements in all-modality models after post-training alignment, we construct a challenging all-modality capability evaluation framework -- eval-anything. All data, models, and code frameworks have been open-sourced for the community. For more details, please refer to https://github.com/PKU-Alignment/align-anything. △ Less

Submitted 30 December, 2024; v1 submitted 20 December, 2024; originally announced December 2024.

Entropy-Regularized Process Reward Model

Authors: Hanning Zhang, Pengcheng Wang, Shizhe Diao, Yong Lin, Rui Pan, Hanze Dong, Dylan Zhang, Pavlo Molchanov, Tong Zhang

Abstract: Large language models (LLMs) have shown promise in performing complex multi-step reasoning, yet they continue to struggle with mathematical reasoning, often making systematic errors. A promising solution is reinforcement learning (RL) guided by reward models, particularly those focusing on process rewards, which score each intermediate step rather than solely evaluating the final outcome. This app… ▽ More Large language models (LLMs) have shown promise in performing complex multi-step reasoning, yet they continue to struggle with mathematical reasoning, often making systematic errors. A promising solution is reinforcement learning (RL) guided by reward models, particularly those focusing on process rewards, which score each intermediate step rather than solely evaluating the final outcome. This approach is more effective at guiding policy models towards correct reasoning trajectories. In this work, we propose an entropy-regularized process reward model (ER-PRM) that integrates KL-regularized Markov Decision Processes (MDP) to balance policy optimization with the need to prevent the policy from shifting too far from its initial distribution. We derive a novel reward construction method based on the theoretical results. Our theoretical analysis shows that we could derive the optimal reward model from the initial policy sampling. Our empirical experiments on the MATH and GSM8K benchmarks demonstrate that ER-PRM consistently outperforms existing process reward models, achieving 1% improvement on GSM8K and 2-3% improvement on MATH under best-of-N evaluation, and more than 1% improvement under RLHF. These results highlight the efficacy of entropy-regularization in enhancing LLMs' reasoning capabilities. △ Less

Submitted 14 December, 2024; originally announced December 2024.

Comments: Preprint

METIS: Fast Quality-Aware RAG Systems with Configuration Adaptation

Authors: Siddhant Ray, Rui Pan, Zhuohan Gu, Kuntai Du, Shaoting Feng, Ganesh Ananthanarayanan, Ravi Netravali, Junchen Jiang

Abstract: RAG (Retrieval Augmented Generation) allows LLMs (large language models) to generate better responses with external knowledge, but using more external knowledge often improves generation quality at the expense of response delay. Prior work either reduces the response delay (through better scheduling of RAG queries) or strives to maximize quality (which involves tuning the RAG workflow), but they f… ▽ More RAG (Retrieval Augmented Generation) allows LLMs (large language models) to generate better responses with external knowledge, but using more external knowledge often improves generation quality at the expense of response delay. Prior work either reduces the response delay (through better scheduling of RAG queries) or strives to maximize quality (which involves tuning the RAG workflow), but they fall short in optimizing the tradeoff between the delay and quality of RAG responses. This paper presents METIS, the first RAG system that jointly schedules queries and adapts the key RAG configurations of each query, such as the number of retrieved text chunks and synthesis methods, in order to balance quality optimization and response delay reduction. Using 4 popular RAG-QA datasets, we show that compared with the state-of-the-art RAG optimization schemes, METIS reduces the generation latency by $1.64-2.54\times$ without sacrificing generation quality. △ Less

Submitted 15 July, 2025; v1 submitted 13 December, 2024; originally announced December 2024.

Comments: 17 pages, 18 figures

SVGBuilder: Component-Based Colored SVG Generation with Text-Guided Autoregressive Transformers

Authors: Zehao Chen, Rong Pan

Abstract: Scalable Vector Graphics (SVG) are essential XML-based formats for versatile graphics, offering resolution independence and scalability. Unlike raster images, SVGs use geometric shapes and support interactivity, animation, and manipulation via CSS and JavaScript. Current SVG generation methods face challenges related to high computational costs and complexity. In contrast, human designers use comp… ▽ More Scalable Vector Graphics (SVG) are essential XML-based formats for versatile graphics, offering resolution independence and scalability. Unlike raster images, SVGs use geometric shapes and support interactivity, animation, and manipulation via CSS and JavaScript. Current SVG generation methods face challenges related to high computational costs and complexity. In contrast, human designers use component-based tools for efficient SVG creation. Inspired by this, SVGBuilder introduces a component-based, autoregressive model for generating high-quality colored SVGs from textual input. It significantly reduces computational overhead and improves efficiency compared to traditional methods. Our model generates SVGs up to 604 times faster than optimization-based approaches. To address the limitations of existing SVG datasets and support our research, we introduce ColorSVG-100K, the first large-scale dataset of colored SVGs, comprising 100,000 graphics. This dataset fills the gap in color information for SVG generation models and enhances diversity in model training. Evaluation against state-of-the-art models demonstrates SVGBuilder's superior performance in practical applications, highlighting its efficiency and quality in generating complex SVG graphics. △ Less

Submitted 12 March, 2025; v1 submitted 13 December, 2024; originally announced December 2024.

Comments: Project: https://svgbuilder.github.io

Journal ref: Proceedings of the AAAI Conference on Artificial Intelligence, 2025, 39(3), 2358-2366

Residual Channel Boosts Contrastive Learning for Radio Frequency Fingerprint Identification

Authors: Rui Pan, Hui Chen, Guanxiong Shen, Hongyang Chen

Abstract: In order to address the issue of limited data samples for the deployment of pre-trained models in unseen environments, this paper proposes a residual channel-based data augmentation strategy for Radio Frequency Fingerprint Identification (RFFI), coupled with a lightweight SimSiam contrastive learning framework. By applying least square (LS) and minimum mean square error (MMSE) channel estimations… ▽ More In order to address the issue of limited data samples for the deployment of pre-trained models in unseen environments, this paper proposes a residual channel-based data augmentation strategy for Radio Frequency Fingerprint Identification (RFFI), coupled with a lightweight SimSiam contrastive learning framework. By applying least square (LS) and minimum mean square error (MMSE) channel estimations followed by equalization, signals with different residual channel effects are generated. These residual channels enable the model to learn more effective representations. Then the pre-trained model is fine-tuned with 1% samples in a novel environment for RFFI. Experimental results demonstrate that our method significantly enhances both feature extraction ability and generalization while requiring fewer samples and less time, making it suitable for practical wireless security applications. △ Less

Submitted 11 December, 2024; originally announced December 2024.

Comments: 5 pages, 4 figures

TDD-Bench Verified: Can LLMs Generate Tests for Issues Before They Get Resolved?

Authors: Toufique Ahmed, Martin Hirzel, Rangeet Pan, Avraham Shinnar, Saurabh Sinha

Abstract: Test-driven development (TDD) is the practice of writing tests first and coding later, and the proponents of TDD expound its numerous benefits. For instance, given an issue on a source code repository, tests can clarify the desired behavior among stake-holders before anyone writes code for the agreed-upon fix. Although there has been a lot of work on automated test generation for the practice "wri… ▽ More Test-driven development (TDD) is the practice of writing tests first and coding later, and the proponents of TDD expound its numerous benefits. For instance, given an issue on a source code repository, tests can clarify the desired behavior among stake-holders before anyone writes code for the agreed-upon fix. Although there has been a lot of work on automated test generation for the practice "write code first, test later", there has been little such automation for TDD. Ideally, tests for TDD should be fail-to-pass (i.e., fail before the issue is resolved and pass after) and have good adequacy with respect to covering the code changed during issue resolution. This paper introduces TDD-Bench Verified, a high-quality benchmark suite of 449 issues mined from real-world GitHub code repositories. The benchmark's evaluation harness runs only relevant tests in isolation for simple yet accurate coverage measurements, and the benchmark's dataset is filtered both by human judges and by execution in the harness. This paper also presents Auto-TDD, an LLM-based solution that takes as input an issue description and a codebase (prior to issue resolution) and returns as output a test that can be used to validate the changes made for resolving the issue. Our evaluation shows that Auto-TDD yields a better fail-to-pass rate than the strongest prior work while also yielding high coverage adequacy. Overall, we hope that this work helps make developers more productive at resolving issues while simultaneously leading to more robust fixes. △ Less

Submitted 3 December, 2024; originally announced December 2024.

Causal Discovery by Interventions via Integer Programming

Authors: Abdelmonem Elrefaey, Rong Pan

Abstract: Causal discovery is essential across various scientific fields to uncover causal structures within data. Traditional methods relying on observational data have limitations due to confounding variables. This paper presents an optimization-based approach using integer programming (IP) to design minimal intervention sets that ensure causal structure identifiability. Our method provides exact and modu… ▽ More Causal discovery is essential across various scientific fields to uncover causal structures within data. Traditional methods relying on observational data have limitations due to confounding variables. This paper presents an optimization-based approach using integer programming (IP) to design minimal intervention sets that ensure causal structure identifiability. Our method provides exact and modular solutions that can be adjusted to different experimental settings and constraints. We demonstrate its effectiveness through comparative analysis across different settings, demonstrating its applicability and robustness. △ Less

Submitted 2 December, 2024; originally announced December 2024.

Leveraging LLM for Automated Ontology Extraction and Knowledge Graph Generation

Authors: Mohammad Sadeq Abolhasani, Rong Pan

Abstract: Extracting relevant and structured knowledge from large, complex technical documents within the Reliability and Maintainability (RAM) domain is labor-intensive and prone to errors. Our work addresses this challenge by presenting OntoKGen, a genuine pipeline for ontology extraction and Knowledge Graph (KG) generation. OntoKGen leverages Large Language Models (LLMs) through an interactive user inter… ▽ More Extracting relevant and structured knowledge from large, complex technical documents within the Reliability and Maintainability (RAM) domain is labor-intensive and prone to errors. Our work addresses this challenge by presenting OntoKGen, a genuine pipeline for ontology extraction and Knowledge Graph (KG) generation. OntoKGen leverages Large Language Models (LLMs) through an interactive user interface guided by our adaptive iterative Chain of Thought (CoT) algorithm to ensure that the ontology extraction process and, thus, KG generation align with user-specific requirements. Although KG generation follows a clear, structured path based on the confirmed ontology, there is no universally correct ontology as it is inherently based on the user's preferences. OntoKGen recommends an ontology grounded in best practices, minimizing user effort and providing valuable insights that may have been overlooked, all while giving the user complete control over the final ontology. Having generated the KG based on the confirmed ontology, OntoKGen enables seamless integration into schemeless, non-relational databases like Neo4j. This integration allows for flexible storage and retrieval of knowledge from diverse, unstructured sources, facilitating advanced querying, analysis, and decision-making. Moreover, the generated KG serves as a robust foundation for future integration into Retrieval Augmented Generation (RAG) systems, offering enhanced capabilities for developing domain-specific intelligent applications. △ Less

Submitted 9 December, 2024; v1 submitted 30 November, 2024; originally announced December 2024.

Marconi: Prefix Caching for the Era of Hybrid LLMs

Authors: Rui Pan, Zhuang Wang, Zhen Jia, Can Karakus, Luca Zancato, Tri Dao, Yida Wang, Ravi Netravali

Abstract: Hybrid models that combine the language modeling capabilities of Attention layers with the efficiency of Recurrent layers (e.g., State Space Models) have gained traction in practically supporting long contexts in Large Language Model serving. Yet, the unique properties of these models complicate the usage of complementary efficiency optimizations such as prefix caching that skip redundant computat… ▽ More Hybrid models that combine the language modeling capabilities of Attention layers with the efficiency of Recurrent layers (e.g., State Space Models) have gained traction in practically supporting long contexts in Large Language Model serving. Yet, the unique properties of these models complicate the usage of complementary efficiency optimizations such as prefix caching that skip redundant computations across requests. Most notably, their use of in-place state updates for recurrent layers precludes rolling back cache entries for partial sequence overlaps, and instead mandates only exact-match cache hits; the effect is a deluge of (large) cache entries per sequence, most of which yield minimal reuse opportunities. We present Marconi, the first system that supports efficient prefix caching with Hybrid LLMs. Key to Marconi are its novel admission and eviction policies that more judiciously assess potential cache entries based not only on recency, but also on (1) forecasts of their reuse likelihood across a taxonomy of different hit scenarios, and (2) the compute savings that hits deliver relative to memory footprints. Across diverse workloads and Hybrid models, Marconi achieves up to 34.4$\times$ higher token hit rates (71.1% or 617 ms lower TTFT) compared to state-of-the-art prefix caching systems. △ Less

Submitted 10 April, 2025; v1 submitted 28 November, 2024; originally announced November 2024.

Comments: MLSys 2025 camera-ready version

Fox-1: Open Small Language Model for Cloud and Edge

Authors: Zijian Hu, Jipeng Zhang, Rui Pan, Zhaozhuo Xu, Shanshan Han, Han Jin, Alay Dilipbhai Shah, Dimitris Stripelis, Yuhang Yao, Salman Avestimehr, Tong Zhang, Chaoyang He

Abstract: We present Fox-1, a series of small language models (SLMs) consisting of Fox-1-1.6B and Fox-1-1.6B-Instruct-v0.1. These models are pre-trained on 3 trillion tokens of web-scraped document data and fine-tuned with 5 billion tokens of instruction-following and multi-turn conversation data. Aiming to improve the pre-training efficiency, Fox-1-1.6B model introduces a novel 3-stage data curriculum acro… ▽ More We present Fox-1, a series of small language models (SLMs) consisting of Fox-1-1.6B and Fox-1-1.6B-Instruct-v0.1. These models are pre-trained on 3 trillion tokens of web-scraped document data and fine-tuned with 5 billion tokens of instruction-following and multi-turn conversation data. Aiming to improve the pre-training efficiency, Fox-1-1.6B model introduces a novel 3-stage data curriculum across all the training data with 2K-8K sequence length. In architecture design, Fox-1 features a deeper layer structure, an expanded vocabulary, and utilizes Grouped Query Attention (GQA), offering a performant and efficient architecture compared to other SLMs. Fox-1 achieves better or on-par performance in various benchmarks compared to StableLM-2-1.6B, Gemma-2B, Qwen1.5-1.8B, and OpenELM1.1B, with competitive inference speed and throughput. The model weights have been released under the Apache 2.0 license, where we aim to promote the democratization of LLMs and make them fully accessible to the whole open-source community. △ Less

Submitted 7 April, 2025; v1 submitted 7 November, 2024; originally announced November 2024.

Comments: Base model is available at https://huggingface.co/tensoropera/Fox-1-1.6B and the instruction-tuned version is available at https://huggingface.co/tensoropera/Fox-1-1.6B-Instruct-v0.1

AlphaTrans: A Neuro-Symbolic Compositional Approach for Repository-Level Code Translation and Validation

Authors: Ali Reza Ibrahimzada, Kaiyao Ke, Mrigank Pawagi, Muhammad Salman Abid, Rangeet Pan, Saurabh Sinha, Reyhaneh Jabbarvand

Abstract: Code translation transforms programs from one programming language (PL) to another. Several rule-based transpilers have been designed to automate code translation between different pairs of PLs. However, the rules can become obsolete as the PLs evolve and cannot generalize to other PLs. Recent studies have explored the automation of code translation using Large Language Models (LLMs). One key obse… ▽ More Code translation transforms programs from one programming language (PL) to another. Several rule-based transpilers have been designed to automate code translation between different pairs of PLs. However, the rules can become obsolete as the PLs evolve and cannot generalize to other PLs. Recent studies have explored the automation of code translation using Large Language Models (LLMs). One key observation is that such techniques may work well for crafted benchmarks but fail to generalize to the scale and complexity of real-world projects with dependencies, custom types, PL-specific features, etc. We propose AlphaTrans, a neuro-symbolic approach to automate repository-level code translation. AlphaTrans translates both source and test code, and employs multiple levels of validation to ensure the translation preserves the functionality of the source program. To break down the problem for LLMs, AlphaTrans leverages program analysis to decompose the program into fragments and translates them in the reverse call order. We leveraged AlphaTrans to translate ten real-world open-source projects consisting of <836, 8575, 2719> classes, methods, and tests. AlphaTrans breaks down these projects into 17874 fragments and translates the entire repository. 96.40% of the translated fragments are syntactically correct, and AlphaTrans validates the translations' runtime behavior and functional correctness for 27.03% and 25.14% of fragments. On average, the integrated translation and validation take 34 hours to translate a project, showing its scalability in practice. For the incorrect translations, AlphaTrans generates a report including existing translation, stack trace, test errors, or assertion failures. We provided these artifacts to two developers to fix the translation bugs in four projects. They were able to fix the issues in 20.1 hours on average and achieve all passing tests. △ Less

Submitted 19 June, 2025; v1 submitted 31 October, 2024; originally announced October 2024.

Comments: Published in FSE 2025

Gaussian Derivative Change-point Detection for Early Warnings of Industrial System Failures

Authors: Hao Zhao, Rong Pan

Abstract: An early warning of future system failure is essential for conducting predictive maintenance and enhancing system availability. This paper introduces a three-step framework for assessing system health to predict imminent system breakdowns. First, the Gaussian Derivative Change-Point Detection (GDCPD) algorithm is proposed for detecting changes in the high-dimensional feature space. GDCPD conducts… ▽ More An early warning of future system failure is essential for conducting predictive maintenance and enhancing system availability. This paper introduces a three-step framework for assessing system health to predict imminent system breakdowns. First, the Gaussian Derivative Change-Point Detection (GDCPD) algorithm is proposed for detecting changes in the high-dimensional feature space. GDCPD conducts a multivariate Change-Point Detection (CPD) by implementing Gaussian derivative processes for identifying change locations on critical system features, as these changes eventually will lead to system failure. To assess the significance of these changes, Weighted Mahalanobis Distance (WMD) is applied in both offline and online analyses. In the offline setting, WMD helps establish a threshold that determines significant system variations, while in the online setting, it facilitates real-time monitoring, issuing alarms for potential future system breakdowns. Utilizing the insights gained from the GDCPD and monitoring scheme, Long Short-Term Memory (LSTM) network is then employed to estimate the Remaining Useful Life (RUL) of the system. The experimental study of a real-world system demonstrates the effectiveness of the proposed methodology in accurately forecasting system failures well before they occur. By integrating CPD with real-time monitoring and RUL prediction, this methodology significantly advances system health monitoring and early warning capabilities. △ Less

Submitted 24 November, 2024; v1 submitted 29 October, 2024; originally announced October 2024.

Bridge-Coder: Unlocking LLMs' Potential to Overcome Language Gaps in Low-Resource Code

Authors: Jipeng Zhang, Jianshu Zhang, Yuanzhe Li, Renjie Pi, Rui Pan, Runtao Liu, Ziqiang Zheng, Tong Zhang

Abstract: Large Language Models (LLMs) demonstrate strong proficiency in generating code for high-resource programming languages (HRPLs) like Python but struggle significantly with low-resource programming languages (LRPLs) such as Racket or D. This performance gap deepens the digital divide, preventing developers using LRPLs from benefiting equally from LLM advancements and reinforcing disparities in innov… ▽ More Large Language Models (LLMs) demonstrate strong proficiency in generating code for high-resource programming languages (HRPLs) like Python but struggle significantly with low-resource programming languages (LRPLs) such as Racket or D. This performance gap deepens the digital divide, preventing developers using LRPLs from benefiting equally from LLM advancements and reinforcing disparities in innovation within underrepresented programming communities. While generating additional training data for LRPLs is promising, it faces two key challenges: manual annotation is labor-intensive and costly, and LLM-generated LRPL code is often of subpar quality. The underlying cause of this issue is the gap between natural language to programming language gap (NL-PL Gap), which is especially pronounced in LRPLs due to limited aligned data. In this work, we introduce a novel approach called Bridge-Coder, which leverages LLMs' intrinsic capabilities to enhance the performance on LRPLs. Our method consists of two key stages. Bridge Generation, where we create high-quality dataset by utilizing LLMs' general knowledge understanding, proficiency in HRPLs, and in-context learning abilities. Then, we apply the Bridged Alignment, which progressively improves the alignment between NL instructions and LRPLs. Experimental results across multiple LRPLs show that Bridge-Coder significantly enhances model performance, demonstrating the effectiveness and generalization of our approach. Furthermore, we offer a detailed analysis of the key components of our method, providing valuable insights for future work aimed at addressing the challenges associated with LRPLs. △ Less

Submitted 24 October, 2024; originally announced October 2024.

Comments: 15 pages, 3 figures

MEC-IP: Efficient Discovery of Markov Equivalent Classes via Integer Programming

Authors: Abdelmonem Elrefaey, Rong Pan

Abstract: This paper presents a novel Integer Programming (IP) approach for discovering the Markov Equivalent Class (MEC) of Bayesian Networks (BNs) through observational data. The MEC-IP algorithm utilizes a unique clique-focusing strategy and Extended Maximal Spanning Graphs (EMSG) to streamline the search for MEC, thus overcoming the computational limitations inherent in other existing algorithms. Our nu… ▽ More This paper presents a novel Integer Programming (IP) approach for discovering the Markov Equivalent Class (MEC) of Bayesian Networks (BNs) through observational data. The MEC-IP algorithm utilizes a unique clique-focusing strategy and Extended Maximal Spanning Graphs (EMSG) to streamline the search for MEC, thus overcoming the computational limitations inherent in other existing algorithms. Our numerical results show that not only a remarkable reduction in computational time is achieved by our algorithm but also an improvement in causal discovery accuracy is seen across diverse datasets. These findings underscore this new algorithm's potential as a powerful tool for researchers and practitioners in causal discovery and BNSL, offering a significant leap forward toward the efficient and accurate analysis of complex data structures. △ Less

Submitted 22 October, 2024; originally announced October 2024.

Optimizing Mixture-of-Experts Inference Time Combining Model Deployment and Communication Scheduling

Authors: Jialong Li, Shreyansh Tripathi, Lakshay Rastogi, Yiming Lei, Rui Pan, Yiting Xia

Abstract: As machine learning models scale in size and complexity, their computational requirements become a significant barrier. Mixture-of-Experts (MoE) models alleviate this issue by selectively activating relevant experts. Despite this, MoE models are hindered by high communication overhead from all-to-all operations, low GPU utilization due to the synchronous communication constraint, and complications… ▽ More As machine learning models scale in size and complexity, their computational requirements become a significant barrier. Mixture-of-Experts (MoE) models alleviate this issue by selectively activating relevant experts. Despite this, MoE models are hindered by high communication overhead from all-to-all operations, low GPU utilization due to the synchronous communication constraint, and complications from heterogeneous GPU environments. This paper presents Aurora, which optimizes both model deployment and all-to-all communication scheduling to address these challenges in MoE inference. Aurora achieves minimal communication times by strategically ordering token transmissions in all-to-all communications. It improves GPU utilization by colocating experts from different models on the same device, avoiding the limitations of synchronous all-to-all communication. We analyze Aurora's optimization strategies theoretically across four common GPU cluster settings: exclusive vs. colocated models on GPUs, and homogeneous vs. heterogeneous GPUs. Aurora provides optimal solutions for three cases, and for the remaining NP-hard scenario, it offers a polynomial-time sub-optimal solution with only a 1.07x degradation from the optimal. Aurora is the first approach to minimize MoE inference time via optimal model deployment and communication scheduling across various scenarios. Evaluations demonstrate that Aurora significantly accelerates inference, achieving speedups of up to 2.38x in homogeneous clusters and 3.54x in heterogeneous environments. Moreover, Aurora enhances GPU utilization by up to 1.5x compared to existing methods. △ Less

Submitted 22 October, 2024; originally announced October 2024.

How Does Data Diversity Shape the Weight Landscape of Neural Networks?

Authors: Yang Ba, Michelle V. Mancenido, Rong Pan

Abstract: To enhance the generalization of machine learning models to unseen data, techniques such as dropout, weight decay ($L_2$ regularization), and noise augmentation are commonly employed. While regularization methods (i.e., dropout and weight decay) are geared toward adjusting model parameters to prevent overfitting, data augmentation increases the diversity of the input training set, a method purport… ▽ More To enhance the generalization of machine learning models to unseen data, techniques such as dropout, weight decay ($L_2$ regularization), and noise augmentation are commonly employed. While regularization methods (i.e., dropout and weight decay) are geared toward adjusting model parameters to prevent overfitting, data augmentation increases the diversity of the input training set, a method purported to improve accuracy and calibration error. In this paper, we investigate the impact of each of these techniques on the parameter space of neural networks, with the goal of understanding how they alter the weight landscape in transfer learning scenarios. To accomplish this, we employ Random Matrix Theory to analyze the eigenvalue distributions of pre-trained models, fine-tuned using these techniques but using different levels of data diversity, for the same downstream tasks. We observe that diverse data influences the weight landscape in a similar fashion as dropout. Additionally, we compare commonly used data augmentation methods with synthetic data created by generative models. We conclude that synthetic data can bring more diversity into real input data, resulting in a better performance on out-of-distribution test instances. △ Less

Submitted 18 October, 2024; originally announced October 2024.

Codellm-Devkit: A Framework for Contextualizing Code LLMs with Program Analysis Insights

Authors: Rahul Krishna, Rangeet Pan, Raju Pavuluri, Srikanth Tamilselvam, Maja Vukovic, Saurabh Sinha

Abstract: Large Language Models for Code (or code LLMs) are increasingly gaining popularity and capabilities, offering a wide array of functionalities such as code completion, code generation, code summarization, test generation, code translation, and more. To leverage code LLMs to their full potential, developers must provide code-specific contextual information to the models. These are typically derived a… ▽ More Large Language Models for Code (or code LLMs) are increasingly gaining popularity and capabilities, offering a wide array of functionalities such as code completion, code generation, code summarization, test generation, code translation, and more. To leverage code LLMs to their full potential, developers must provide code-specific contextual information to the models. These are typically derived and distilled using program analysis tools. However, there exists a significant gap--these static analysis tools are often language-specific and come with a steep learning curve, making their effective use challenging. These tools are tailored to specific program languages, requiring developers to learn and manage multiple tools to cover various aspects of the their code base. Moreover, the complexity of configuring and integrating these tools into the existing development environments add an additional layer of difficulty. This challenge limits the potential benefits that could be gained from more widespread and effective use of static analysis in conjunction with LLMs. To address this challenge, we present codellm-devkit (hereafter, `CLDK'), an open-source library that significantly simplifies the process of performing program analysis at various levels of granularity for different programming languages to support code LLM use cases. As a Python library, CLDK offers developers an intuitive and user-friendly interface, making it incredibly easy to provide rich program analysis context to code LLMs. With this library, developers can effortlessly integrate detailed, code-specific insights that enhance the operational efficiency and effectiveness of LLMs in coding tasks. CLDK is available as an open-source library at https://github.com/IBM/codellm-devkit. △ Less

Submitted 16 October, 2024; originally announced October 2024.

Fill In The Gaps: Model Calibration and Generalization with Synthetic Data

Authors: Yang Ba, Michelle V. Mancenido, Rong Pan

Abstract: As machine learning models continue to swiftly advance, calibrating their performance has become a major concern prior to practical and widespread implementation. Most existing calibration methods often negatively impact model accuracy due to the lack of diversity of validation data, resulting in reduced generalizability. To address this, we propose a calibration method that incorporates synthetic… ▽ More As machine learning models continue to swiftly advance, calibrating their performance has become a major concern prior to practical and widespread implementation. Most existing calibration methods often negatively impact model accuracy due to the lack of diversity of validation data, resulting in reduced generalizability. To address this, we propose a calibration method that incorporates synthetic data without compromising accuracy. We derive the expected calibration error (ECE) bound using the Probably Approximately Correct (PAC) learning framework. Large language models (LLMs), known for their ability to mimic real data and generate text with mixed class labels, are utilized as a synthetic data generation strategy to lower the ECE bound and improve model accuracy on real test data. Additionally, we propose data generation mechanisms for efficient calibration. Testing our method on four different natural language processing tasks, we observed an average up to 34\% increase in accuracy and 33\% decrease in ECE. △ Less

Submitted 7 October, 2024; originally announced October 2024.

Comments: Accepted to EMNLP 2024 Main Conference (Long paper)

Exploring Foundation Models in Remote Sensing Image Change Detection: A Comprehensive Survey

Authors: Zihan Yu, Tianxiao Li, Yuxin Zhu, Rongze Pan

Abstract: Change detection, as an important and widely applied technique in the field of remote sensing, aims to analyze changes in surface areas over time and has broad applications in areas such as environmental monitoring, urban development, and land use analysis.In recent years, deep learning, especially the development of foundation models, has provided more powerful solutions for feature extraction an… ▽ More Change detection, as an important and widely applied technique in the field of remote sensing, aims to analyze changes in surface areas over time and has broad applications in areas such as environmental monitoring, urban development, and land use analysis.In recent years, deep learning, especially the development of foundation models, has provided more powerful solutions for feature extraction and data fusion, effectively addressing these complexities. This paper systematically reviews the latest advancements in the field of change detection, with a focus on the application of foundation models in remote sensing tasks. △ Less

Submitted 10 October, 2024; originally announced October 2024.

Comments: 14 pages